API Reference

Beta

Agent Data

Copy Markdown

Open in Claude

Open in ChatGPT

Open in Cursor

---

Copy Markdown

View as Markdown

Aggregate Agent Data

client.beta.agentData.aggregate(AgentDataAggregateParams { deployment_name, organization_id, project_id, 9 more } params, RequestOptionsoptions?): PaginatedCursorPost<AgentDataAggregateResponse { group_key, count, first_item } >

POST/api/v1/beta/agent-data/:aggregate

Aggregate agent data with grouping and optional counting/first item retrieval.

ParametersExpand Collapse

params: AgentDataAggregateParams { deployment_name, organization_id, project_id, 9 more }

deployment_name: string

Body param: The agent deployment’s name to aggregate data for

organization_id?: string | null

Query param

formatuuid

project_id?: string | null

Query param

formatuuid

collection?: string

Body param: The logical agent data collection to aggregate data for

count?: boolean | null

Body param: Whether to count the number of items in each group

filter?: Record<string, Filter> | null

Body param: A filter object or expression that filters resources listed in the response.

eq?: number | string | (string & {}) | null

One of the following:

number

string

(string & {})

excludes?: Array<number | string | (string & {}) | null>

One of the following:

number

string

(string & {})

gt?: number | string | (string & {}) | null

One of the following:

number

string

(string & {})

gte?: number | string | (string & {}) | null

One of the following:

number

string

(string & {})

includes?: Array<number | string | (string & {}) | null>

One of the following:

number

string

(string & {})

lt?: number | string | (string & {}) | null

One of the following:

number

string

(string & {})

lte?: number | string | (string & {}) | null

One of the following:

number

string

(string & {})

ne?: number | string | (string & {}) | null

One of the following:

number

string

(string & {})

first?: boolean | null

Body param: Whether to return the first item in each group (Sorted by created_at)

group_by?: Array<string> | null

Body param: The fields to group by. If empty, the entire dataset is grouped on. e.g. if left out, can be used for simple count operations

offset?: number | null

Body param: The offset to start from. If not provided, the first page is returned

maximum1000

minimum0

order_by?: string | null

Body param: A comma-separated list of fields to order by, sorted in ascending order. Use ‘field_name desc’ to specify descending order.

page_size?: number | null

Body param: The maximum number of items to return. The service may return fewer than this value. If unspecified, a default page size will be used. The maximum value is typically 1000; values above this will be coerced to the maximum.

page_token?: string | null

Body param: A page token, received from a previous list call. Provide this to retrieve the subsequent page.

ReturnsExpand Collapse

AgentDataAggregateResponse { group_key, count, first_item }

API Result for a single group in the aggregate response

group_key: Record<string, unknown>

count?: number | null

first_item?: Record<string, unknown> | null

Aggregate Agent Data

TypeScript

HTTP

TypeScript

Python

Go

Java

CLI Tool

import LlamaCloud from '@llamaindex/llama-cloud';

const client = new LlamaCloud({
  apiKey: process.env['LLAMA_CLOUD_API_KEY'], // This is the default and can be omitted
});

// Automatically fetches more pages as needed.
for await (const agentDataAggregateResponse of client.beta.agentData.aggregate({
  deployment_name: 'deployment_name',
})) {
  console.log(agentDataAggregateResponse.group_key);
}

200 example

{
  "items": [
    {
      "group_key": {
        "foo": "bar"
      },
      "count": 0,
      "first_item": {
        "foo": "bar"
      }
    }
  ],
  "next_page_token": "next_page_token",
  "total_size": 0
}

Returns Examples

200 example

{
  "items": [
    {
      "group_key": {
        "foo": "bar"
      },
      "count": 0,
      "first_item": {
        "foo": "bar"
      }
    }
  ],
  "next_page_token": "next_page_token",
  "total_size": 0
}